ozbolt/luscenje_struktur
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Adding changable pc tag (when it is c and not pc)

Browse Source
This commit is contained in:
Ozbolt Menegatti 2019-02-12 12:08:30 +01:00
parent c1e85255c7
commit 2d373ab477
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
wani.py
View File

@ -746,10 +746,10 @@ def load_files(args):
do_msd_translate = not args.no_msd_translate
for fname in filenames:
yield load_tei_file(fname, skip_id_check, do_msd_translate)
yield load_tei_file(fname, skip_id_check, do_msd_translate, args.pc_tag)
def load_tei_file(filename, skip_id_check, do_msd_translate):
def load_tei_file(filename, skip_id_check, do_msd_translate, pc_tag):
logging.info("LOADING FILE: {}".format(filename))
with open(filename, 'r') as fp:
@ -760,7 +760,7 @@ def load_tei_file(filename, skip_id_check, do_msd_translate):
words = {}
for w in et.iter("w"):
words[w.get('id')] = Word(w, do_msd_translate)
for pc in et.iter("pc"):
for pc in et.iter(pc_tag):
words[pc.get('id')] = Word.pcWord(pc, do_msd_translate)
for l in et.iter("link"):
@ -973,6 +973,8 @@ if __name__ == '__main__':
parser.add_argument('--verbose', help='Enable verbose output to stderr', choices=["warning", "info", "debug"], default="info")
parser.add_argument('--multiple-output', help='Generate one output for each syntactic structure', action='store_true')
parser.add_argument('--pc-tag', help='Tag for separators, usually pc or c', default="pc")
args = parser.parse_args()
logging.basicConfig(stream=sys.stderr, level=args.verbose.upper())